Target for campaign success: an empirical analysis of equity crowdfunding in Japan

Abstract

This study investigates campaign success in equity crowdfunding, using campaigns listed on a leading Japanese equity crowdfunding platform with an all-or-nothing rule. We examine how campaign success depends on campaign- and firm-specific characteristics, including the campaign target amount. We find that patenting positively affects campaign success and the campaign target amount. The results indicate that innovative firms have a higher probability of campaign success, while they set a higher target amount. We also provide evidence that campaigns that have already provided services and products have a lower probability of success, although not always significant, and firms tend to set a lower target amount. Moreover, campaigns launched by venture capital-backed firms are more likely to succeed than others. Furthermore, campaigns launched by firms eligible for the Angel Tax System, introduced in Japan as a tax incentive for investment in young and small firms, have a higher tendency to succeed in equity crowdfunding. Our findings suggest that the Angel Tax System attract more crowd investors and such tax incentives help raise funds on equity crowdfunding platforms for new businesses.

1 Introduction

Equity crowdfunding attracts a substantial number of investors and funds from many countries and regions. It provides a great opportunity to increase financial inclusion of classes of entrepreneurs facing severe challenges in accessing financing (Butticè & Vismara, 2022). Numerous scholars have addressed equity crowdfunding, following an increase in its recognition since the 2010s.¹ Previous studies from the entrepreneurial perspective have provided ample evidence of equity crowdfunding in some economies, such as Australia (Ahlers et al., 2015), France (Cumming et al., 2022), Germany (Blaseg et al., 2021; Block et al., 2018; Hornuf & Schwienbacher, 2018), the UK (Coakley et al., 2022; Cumming et al., 2021; Kleinert et al., 2020; Ralcheva & Roosenboom, 2020; Shafi, 2021; Vismara, 2016, 2018a, 2019; Vulkan et al., 2016), and the US (Borchers & Dunham, 2022; Cummings et al., 2020; Mamonov & Malaga, 2019).² However, while many studies have focused on equity crowdfunding mainly in European economies, there is a paucity of research on equity crowdfunding in Asian economies, including Japan.

Equity crowdfunding is expected to provide funds to entrepreneurs, including high-tech start-ups, especially in Japan, where private equity capital, including venture capital (VC), is underdeveloped. According to Global Entrepreneurship Monitor data, Japan has the lowest level of investment by individuals in new businesses among developed economies (Honjo, 2015; Honjo & Nakamura, 2020).³ To promote funds from angel investors (business angels), the Japanese government introduced Angel Tax System in 1997 as a tax incentive for investment in young and small firms. After April 2020, equity crowdfunding platform firms certified by the Ministry of Economy, Trade and Industry, Japan, could issue whether an equity crowdfunding campaign is eligible for the Angel Tax System. Tax incentives are often designed to encourage seed investment in early-stage firms (Cumming et al., 2019; Vismara, 2016, 2018a, 2019). The Angel Tax System is expected to encourage crowd investors to invest in equity crowdfunding campaigns because of a deduction from gross income or capital gains. Such tax incentives allow investors to make high-risk investments. However, equity crowdfunding in Japan has specific regulations with limitations to the amount of funding. Specifically, an investor is permitted to invest no more than 500 thousand JPY per firm each year, and a firm cannot totally raise 100 million JPY or more per year. By examining equity crowdfunding in Japan, we can better understand how public policies and regulations affect entrepreneurial finance on equity crowdfunding platforms.

Previous studies have focused on equity crowdfunding participation measured by the number of investors and the funding amount (Ahlers et al., 2015; Block et al., 2018; Lukkarinen et al., 2016). Several scholars have extensively examined the probability of campaign success (Ahlers et al., 2015; Cumming et al., 2022; Hornuf et al., 2022; Shafi, 2021; Vismara, 2019). These studies provide ample evidence on the determinants of campaign success, given that the amount pledged in equity crowdfunding depends on campaign-, entrepreneur-, and firm-specific characteristics. In the literature, campaign success is defined as a case in which the amount pledged exceeds 100% or more of the campaign target amount (Colombo et al., 2015; Piva & Rossi-Lamastra, 2018). Campaign success is pivotal for entrepreneurs and crowd investors (crowd funders), especially in the case of an “all-or-nothing” rule, which means that campaigns are funded only if the amount pledged reaches a publicly declared threshold (i.e., campaign target amount). This is because, without campaign success, entrepreneurs do not obtain any funds pledged and crowd investors do not obtain any rewards in equity crowdfunding, even if the amount pledged by a campaign is high. Meanwhile, the campaign target amount is considered to be associated with campaign-specific characteristics. While some firms seek a small amount of equity financing, those devoted to research and development (R&D) require a large amount of R&D investment. More importantly, the probability of campaign success depends not only on the amount pledged but also on the campaign target amount. Firms have an incentive for setting a lower target amount to ensure campaign success (Nitani et al., 2019). Therefore, some firms may set a lower target amount if they are willing to prioritize advertising their developing services and products on equity crowdfunding platforms. Stringent regulations in equity crowdfunding may also promote such a lower setting of a target amount. Indeed, several scholars have found that the campaign target amount has a negative effect on campaign success (Cumming et al., 2021; Piva & Rossi-Lamastra, 2018; Ralcheva & Roosenboom, 2020; Vulkan et al., 2016). These findings suggest that the campaign target amount is strategically determined by entrepreneurs who have an incentive to increase the probability of campaign success by setting a lower target amount. In this respect, scholars should pay more attention to the campaign target amount on equity crowdfunding platforms.

Furthermore, campaigns failing by a small margin are rare, implying that campaigns on the borderline between success and failure are more likely to succeed at the last moment because crowd investors work together to achieve campaign target amounts. At the same time, attractive projects achieve campaign target amounts faster in equity crowdfunding. In other words, the time required for the campaign to succeed may be related to campaign attractiveness. As shown later, most successful campaigns achieve the target amount within 24 h. In this respect, the time taken to achieve campaign success (i.e., time to campaign success) may reflect the level of campaign attractiveness more correctly than the probability of campaign success. Nevertheless, while most studies have focused only on the probability of campaign success, only a few have examined the time to campaign success (Nitani et al., 2019). Further investigation would help to identify the determinants of campaign success on equity crowdfunding platforms.

This study investigates campaign success in equity crowdfunding, using campaigns listed on FUNDINNO. FUNDINNO is a leading Japanese equity crowdfunding platform, which can be regarded as an entrepreneur-led platform with an all-or-nothing rule.⁴ As previously mentioned, campaign success is pivotal for entrepreneurs and crowd investors in the case of the all-or-nothing rule. We examine how campaign success depends on campaign- and firm-specific characteristics, including the campaign target amount. We find that patenting positively affects campaign success and the campaign target amount. The results indicate that innovative firms, measured by patenting, have a higher probability of campaign success, while they set a higher target amount. We also provide evidence that campaigns that have already provided services and products have a lower probability of success, although not always significant, and firms tend to set a lower target amount, suggesting that these firms aim to investigate the market and advertise their services and products rather than raise funds on equity crowdfunding platforms. Moreover, campaigns launched by VC-backed firms are more likely to succeed than others. Furthermore, campaigns launched by firms eligible for the Angel Tax System have a higher tendency to succeed in equity crowdfunding.

The unique features and contributions of this study are fourfold. First, we examine campaign success in equity crowdfunding by identifying the factors affecting the campaign target amount. Our results reveal that patenting positively affects not only campaign success but also the campaign target amount, suggesting that technology-based campaigns are more likely to achieve campaign success, even if entrepreneurs require large funds. Second, we provide novel insights into the advertising and promotion effect of campaigns on equity crowdfunding platforms by identifying a negative relationship with the campaign target amount. More specifically, the campaign target amount, in addition to the probability of campaign success, is lower when services and products have already been developed. This suggests that some campaigns aim to investigate the market and advertise their services and products rather than raise funds on equity crowdfunding platforms. Third, we shed light on evidence on the time to the campaign target amount, that is, the time to campaign success, which may represent the level of campaign attractiveness. We also find that over three-quarters of campaigns launched by firms with patenting or eligibility for the Angel Tax System and those launched by VC-backed firms achieve the campaign target amount within 24 h. Our results indicate that these types of campaigns are more likely to attract crowd investors. Fourth and finally, we provide various pieces of evidence on equity crowdfunding using the case of Japan. New financing techniques, including equity crowdfunding, are often regulated, dependent on the country context. How equity crowdfunding works as a financial means may be driven by public policies and regulations. More importantly, our findings suggest that the Angel Tax System helps sustain entrepreneurial finance in Japanese equity crowdfunding, which can be regarded as a natural experiment for tax policy. Furthermore, stringent regulations due to strong investor protection are imposed on equity crowdfunding platforms in Japan, which may lead to distorted purposes, such as advertising and promotion. Our findings would open the way for debating policies and regulations for financial alternatives to seeding entrepreneurship and innovation.

The remainder of this paper is organized as follows. Section 2 introduces the research background and reviews the literature. Section 3 presents hypotheses about campaign success in equity crowdfunding. Section 4 introduces the data used in the study. Section 5 presents our estimation results. The last section provides concluding remarks.

2 Research background

2.1 Literature on equity crowdfunding

Start-up firms, especially high-tech start-ups, often face financing difficulties because they have highly uncertain prospects, lack tangible assets that can be used as collateral, and face severe information problems (Bernstein et al., 2017; Hall & Lerner, 2010). Equity crowdfunding is expected to play a supporting role in reducing early-stage funding gaps (Brüntje & Gaijda, 2016; Mochkabadi & Volkmann, 2020). It involves the benefits of fast access to a large pool of crowd investors and feedback from the equity crowdfunding market (Blaseg et al., 2021). As seen in the UK, the large financial flows to entrepreneurs via equity crowdfunding have been largely incremental to the traditional sources of early-stage entrepreneurial finance (Estrin et al., 2018).

Studies on equity crowdfunding often consider five perspectives: (i) capital markets, (ii) entrepreneurs, (iii) institutions, (iv) investors, and (v) platforms (Mochkabadi & Volkmann, 2020).⁵ Among these, the entrepreneurial perspective has recently become more relevant. In addition, there are four types of crowdfunding business models: (i) donation-based, (ii) reward-based, (iii) lending, and (iv) equity (Cumming & Hornuf, 2018; Vulkan et al., 2016).⁶ Equity crowdfunding has become an alternative to traditional funding sources, including VC and angel investors (Ralcheva & Roosenboom, 2020). Equity financing is often argued to have advantages over debt financing in R&D investment (Carpenter & Petersen, 2002; Hall, 2002). In particular, high-tech start-ups, including entrepreneurial inventors, require equity financing rather than debt financing (Colombo & Grilli, 2007; Hall, 2010; Honjo, 2021a, 2021b). This is due to the business characteristics associated with high uncertainty and information asymmetry with external suppliers of capital. Young and small firms may rely on equity crowdfunding because of the lack of credit history and track records. Hence, equity crowdfunding is expected to become a popular financing choice for young and small firms (Ralcheva & Roosenboom, 2020).

Table 1 reports some empirical studies that target campaign success in equity crowdfunding from the entrepreneurial perspective.⁷ While many studies have examined equity crowdfunding platforms in European economies, such as the UK, there is a paucity of research on equity crowdfunding in Asian economies, such as Japan. These studies examine the probability of campaign success using data from equity crowdfunding platforms. Crowd investors seek to make investment decisions, considering the probability of campaign success, especially on equity crowdfunding platforms with the all-or-nothing rule. Therefore, the determinants of campaign success are valuable for entrepreneurs and crowd investors, in addition to crowdfunding platform firms, to improve their performance.

Table 1 Empirical studies on campaign success in equity crowdfunding

Furthermore, only a few studies have focused on time to campaign success (Nitani et al., 2019). As is often seen, campaigns on the borderline between success and failure are more likely to succeed at the last moment. In this respect, examining the time to campaign success in equity crowdfunding is useful not only for the urgent financing of firms, but also for a better understanding of what types of campaigns attract more crowd investors.

2.2 Determinants of campaign success

Crowd investors are willing to choose high-quality projects because they can secure returns on investments by participating in successful campaigns. The three dimensions for assessing project quality—human, social, and intellectual capital—have been addressed in the literature (Ahlers et al., 2015). However, crowd investors do not necessarily have sufficient information about project quality; that is, information asymmetries exist between entrepreneurs and crowd investors. Presumably, information asymmetries in equity crowdfunding are more severe in entrepreneurial finance markets (Vismara, 2018b). This may be caused by crowd investors’ lack of experience and capability to evaluate investment opportunities (Ahlers et al., 2015). For entrepreneurs, there are potential costs associated with equity crowdfunding, such as the early public disclosure of entrepreneurial activities, communication costs with a large pool of crowd investors, and equity dilution associated with the discouragement of future investors (Blaseg et al., 2021). Under the information asymmetries, crowd investors also face adverse selection (Walthoff-Borm et al., 2018a, 2018b). With adverse selection, entrepreneurs with high-quality projects find it difficult to obtain high valuations (Meoli et al., 2022). Such entrepreneurs may avoid equity financing because of loss of ownership and opportunity costs.⁸

Generally, crowd investors have imperfect information when evaluating project quality, partly because entrepreneurs provide fragmented information through their project descriptions (Kleinert et al., 2020). Unlike professional investors, such as VC, crowd investors lack the experience and capability to evaluate investment opportunities (Ahlers et al., 2015). Rather, equity crowdfunding can be built on the wisdom of a crowd (Schwienbacher & Larralde, 2012). While professional investors seek detailed due diligence in selecting firms, crowd investors on equity crowdfunding platforms have less incentive to devote substantial resources to the due diligence process (Vismara, 2019). Owing to limited experience in selecting investments, crowd investors’ decisions are susceptible to professional investors’ decisions to invest in campaigns.

Campaign success in equity crowdfunding depends on how campaigns attract crowd investors and promote investment decisions. If an entrepreneur provides valuable information that attracts crowd investors, such a campaign will increase the probability of success in equity crowdfunding. Some scholars have addressed the signaling effect of equity crowdfunding (Ahlers et al., 2015; Bapna, 2019; Block et al., 2018; Piva & Rossi-Lamastra, 2018; Vismara, 2016). In equity crowdfunding platforms, information asymmetries regarding project quality exist between entrepreneurs and crowd investors. Under these conditions, signaling associated with project quality plays a significant role in raising funds and determining campaign success in equity crowdfunding. To raise the necessary funding for business development, entrepreneurs voluntarily disclose business information, such as qualitative business descriptions of business models, to crowd investors through online platforms or portals (Johan & Zhang, 2021). Under information asymmetries between entrepreneurs and crowd investors, credible signaling plays a significant role in raising funds.

According to the signaling theory, some campaign-specific characteristics may determine campaign success. Specifically, intellectual capital increases the probability of campaign success because it represents project quality (Ahlers et al., 2015). More importantly, third parties can certify project quality to outside investors, and prior financing positively affects campaign success (Kleinert et al., 2020).⁹ Indeed, several scholars have found that successful campaigns in equity crowdfunding attract professional investors, such as VC (Butticè et al., 2020; Hornuf et al., 2018). Investors with a public profile increase the offer’s appeal among early investors, who, in turn, attract late investors (Vismara, 2018a). Many firms raise funds through angel investors, VC, crowdfunding, and grants prior to equity crowdfunding campaigns. In particular, the presence of professional investors, such as VC, may signal project quality to crowd investors; thus, the types of investors participating in campaigns—in other words, current equity capital—may influence campaign success in equity crowdfunding.

Furthermore, other studies have examined policy support for investments through equity crowdfunding. Tax incentives are beneficial for investors. In practice, previous studies examined the impact of policy support on campaign success using tax incentives in the UK, specifically the Seed Enterprise Investment Scheme, designed to encourage seed investments in early-stage firms (Cumming et al., 2019; Vismara, 2016, 2018a, 2019). However, the results did not reveal a significant effect of tax incentives on the probability of campaign success. Further investigation is required to evaluate the policy support for the promotion of seed investment.

2.3 Campaign target amount and purpose of equity crowdfunding

The campaign target amount indicates entrepreneurs’ demand for financing on equity crowdfunding platforms, whereas the total amount raised in equity crowdfunding represents the supply of financing from crowd investors. Some scholars have examined the impact of the campaign target amount on the total amount raised (Cumming et al., 2021; Rossi et al., 2021; Vismara, 2016). Moreover, the campaign target amount, as well as the campaign duration, is controllable by entrepreneurs and firms, while the total amount raised is uncontrollable. The campaign target amount is more likely to reflect the firm strategy on equity crowdfunding platforms.

Several scholars have found a negative relationship between the campaign target amount and campaign success (Cumming et al., 2021; Piva & Rossi-Lamastra, 2018; Ralcheva & Roosenboom, 2020; Vulkan et al., 2016).¹⁰ These findings suggest that entrepreneurs are incentivized to set a lower target amount and increase the probability of campaign success. This incentive becomes stronger, especially in the case of the all-or-nothing rule, because entrepreneurs do not obtain any funds pledged without campaign success. In this respect, entrepreneurs strategically determine the campaign target amount. However, to the best of our knowledge, only a few studies have estimated the determinants of the campaign target amount or considered endogeneity (Rossi et al., 2021). Such an investigation of the campaign target amount, including campaign success, could provide a comprehensive understanding of equity crowdfunding success.

Firms and project teams may seek to convey business value and advertise funding campaigns to potential investors (Johan & Zhang, 2020). Although equity crowdfunding differs from purchase-based crowdfunding, firms can identify advertising opportunities on such platforms. Start-up firms that rely on equity crowdfunding may lack a reputation in the financial, labor, and product markets; therefore, entrepreneurs are incentivized to advertise services and products, including their firms themselves, on equity crowdfunding platforms. Crowd investors may provide other non-financial benefits, such as feedback and market validation (Walthoff-Borm et al., 2018a, 2018b). They may provide valuable feedback and become brand ambassadors for the firm and its products (Schwienbacher & Larralde, 2012). Such advertising and promotion effects in equity crowdfunding may be associated with the campaign target amount.

2.4 The Japanese context

New financial alternatives to seeding entrepreneurship, including crowdfunding, have been developed in many economies (Bruton et al., 2015). This is partly because bank financing has become less available since the global financial crisis of the late 2000s (Clayton et al., 2018; Harrison, 2013). However, bank debt financing still plays a central role in some economies, such as Japan, which has established a bank-centered financial system (Pinkowitz & Williamson, 2001; Sasaki & Suzuki, 2019; Weinstein & Yafeh, 1998). It can be manifested whether equity crowdfunding is required for entrepreneurs, by focusing on bank-centered financial systems, such as Japan, rather than market-based financial systems.

In Japan, various programs facilitate the flow of funds for lending to small and medium enterprises (SMEs). Regional banks, shinkin banks (credit associations), and credit cooperatives have been established to fund local businesses. Regarding business start-ups, the Japan Financial Corporation (JFC), a government-affiliated financial institution, has loan programs to promote these activities by providing a stable supply of long-term loans. In particular, public policies have been implemented to enhance debt financing for SMEs, including start-up firms, whenever the Japanese economy stagnates.

Figure 1 displays the number of campaigns and total amount on equity crowdfunding platforms in Japan. Equity crowdfunding has been developing in Japan in recent years. As of December 2021, five equity crowdfunding platform firms (FUNDINNO, CAMPFIRE Startups, Unicorn, Angel Navi, and Ecrowd), registered as Type-I Small-Amount Electronic Public Offering Service Providers and members of the Japan Securities Dealers Association (JSDA), have operated equity crowdfunding platforms. As of August 2021, approximately 78% of equity crowdfunding campaigns in Japan were listed on FUNDINNO, a leading equity crowdfunding platform.¹¹ Because of its indirect network effects, crowdfunding is often dominated by a single platform (Agrawal et al., 2014).

Fig. 1

Number of campaigns and total amount pledged on equity crowdfunding platforms in Japan. Source: JSDA website https://www.jsda.or.jp/shiryoshitsu/toukei/kabucrowdfunding/index.html?_ga=2.144795764.1107716731.1647926998-1350480052.1647926998 [accessed on March 22, 2022]. Note The number of campaigns indicates the number of issued stocks and granted stock options on FUNDINNO and other equity crowdfunding platforms in Japan, regardless of campaign success. The total amount indicates the total amount (million JPY) pledged through issued stocks and stock options on FUNDINNO and other equity crowdfunding platforms in Japan. The figures are collected every three months, but two months only for Q3 2021 (i.e., from July 2021 to August 2021)

Equity crowdfunding in Japan is strictly regulated. Specifically, an investor is permitted to invest no more than 500 thousand JPY per firm each year, and a firm cannot totally raise 100 million JPY or more per year following the Financial Instrument and Exchange Act. In the past, a firm was permitted to raise a maximum aggregate amount of 1.07 million USD yearly in the US, similar to Japan’s upper limit. However, funding regulations in the US were partially amended in March 2021, and a firm was allowed to raise up to 5 million USD. Moreover, the upper limit is not available for firms and most investors in the UK. In this respect, we realize that Japan’s equity crowdfunding regulations are much more stringent.¹² Such stringent regulations in Japan are due to strong investor protection in equity crowdfunding.

Furthermore, the Angel Tax System in Japan has been applied to equity crowdfunding platforms operated by Type-I Small-Amount Electronic Public Offering Service Providers. The Angel Tax System consists of two types of tax benefits: Types A and B. Type A is a deduction from gross income (deductions calculated by the amount of investment in target firms minus 2 thousand JPY) for the year. It can be applied to investment in firms less than three years old (five years after the tax reform in April 2020). Type B is a deduction from capital gains (deduction of the amount invested in target firms from other capital gains) for the year. It can be applied to investment in firms less than 10 years old (Small and Medium Enterprise Agency, 2017). After April 2020, equity crowdfunding platform firms certified by the Ministry of Economy, Trade and Industry, Japan, can issue whether an equity crowdfunding campaign is eligible for the Angel Tax System.¹³ Such tax incentives are expected to promote investment in equity crowdfunding campaigns because of a deduction from gross income or capital gains.

3 Hypotheses development

As high-quality projects are more likely to attract crowd investors, signaling associated with project quality may affect campaign success under information asymmetries between entrepreneurs and crowd investors. Crowd investors expect patent-related campaigns to exhibit higher levels of skill and technology. From the signaling theory perspective, patents signal project quality to outside investors (Walthoff‐Borm et al., 2018a). Intellectual property, such as patents, can be viewed as a credible signal that conveys information about a firm’s innovation capabilities to crowd investors (Ralcheva & Roosenboom, 2020). Some studies have found that firms that use equity crowdfunding have a higher rate of patent applications (Walthoff‐Borm et al. 2018a, 2018b).

Meanwhile, previous studies emphasized the impact of patents on campaign success, especially for campaigns on investor-led platforms rather than entrepreneur-led platforms in the UK (Rossi et al., 2021). However, entrepreneurs pursuing technological development may seek entrepreneur-led platforms in economies with underdeveloped private equity capital, such as Japan. This is also because entrepreneurs lack options for investor-led platforms. In addition, the upper amount of funding is limited even for technology-based campaigns that attract more crowd investors because of their high ability and growth potential; thus, these campaigns have a higher probability of achieving the target amount. Moreover, research campaigns may attract crowd investors if equity crowdfunding is recognized as supporting technological development, including academic activities. Therefore, campaigns proposed by entrepreneurs pursuing technological development are more likely to succeed even in entrepreneur-led platforms. Hence, we test the following hypothesis:

H1

Campaigns launched by firms with patent applications and registrations are more likely to succeed in equity crowdfunding than others.

Entrepreneurs expect financial and non-financial benefits from equity crowdfunding, unlike from other financing sources, such as bank loans. Campaigns already providing their services and products may expect non-financial benefits, such as feedback and market validation, on equity crowdfunding platforms (Walthoff-Borm et al., 2018a, 2018b). Some entrepreneurs may mainly show off their campaigns and tend to set a lower target amount.¹⁴ Others may seek advertisements for their services and products, including test marketing, rather than raise funds. Moreover, entrepreneurs may not expect significant financing because equity crowdfunding is limited to the upper amount that can be raised in Japanese equity crowdfunding. Under these regulations, campaigns with the development of services and products may pursue advertising and promotion rather than crowdfunding. However, crowd investors have less incentive to help promote the developed services and products by providing additional funds. Thus, we consider that advertising and promotion effects in equity crowdfunding result in a lower probability of campaign success. We test whether the development of services and products is associated with campaign success.

H2

Campaigns associated with the development of services and products are less likely to succeed in equity crowdfunding than others.

The presence of professional investors, such as VC, may give crowd investors a sense of security for monetary returns, as professional investors invest in equity crowdfunding to realize monetary returns (Vismara, 2019). VC typically uses the most selective and reliable screening processes (Kleinert et al., 2020; Mason & Harrison, 2002; Van Osnabrugge, 2000). Crowd investors may free-ride on the investment decisions of such professional backers in an information cascade fashion (Signori & Vismara, 2018; Vismara, 2018a). In addition, affiliations with VC are expected to improve firm performance; therefore, they may increase the chances of obtaining follow-on funding (Kleinert et al., 2020). Thus, it is plausible that VC participation prior to equity crowdfunding significantly affects the probability of campaign success. However, it is unclear whether the presence of professional investors, such as VC, attracts crowd investors, partly because VC participation in the early stages prevents crowd investors from obtaining control rights. Some crowd investors are unwilling to participate in campaigns controlled by professional investors. To examine whether the presence of professional investors, such as VC, attracts crowd investors on equity crowdfunding platforms, we test the following hypothesis:

H3

Campaigns launched by professional investor-backed firms, such as VC-backed firms, are more likely to succeed in equity crowdfunding than others.

Furthermore, we examine whether policies introduced in equity crowdfunding in Japan—specifically the Angel Tax System—effectively increase the probability of campaign success. Previous studies emphasize that tax incentives are designed to encourage seed investment in early-stage firms (Cumming et al., 2019; Vismara, 2016, 2018a, 2019). Such tax incentives allow investors to make high-risk investments. As discussed in Sect. 2.4, the Angel Tax System has been applied to equity crowdfunding in Japan. Whereas the Angel Tax System does not signal the quality of campaigns to crowd investors, campaigns launched by firms eligible for the Angel Tax System create investors’ incentives for seed investment because they present a deduction from their gross income or capital gains. Therefore, the Angel Tax System is expected to encourage seed investments by crowd investors. In particular, the Angel Tax System attracts more investors, including angel investors. Some studies also emphasize the role of campaign backers in the early stages through the lens of herding and informational cascades (Petit & Wirtz, 2022; Vismara, 2018a). Although crowd investors do not have the same experience as professional investors, they may make a strategic investment (Wallmeroth, 2019). They may crowd campaigns launched by firms eligible for the Angel Tax System. More importantly, the participation of many crowd investors is required to achieve campaign success because individual investment is limited to a small amount (i.e., no more than 500 thousand JPY per firm each year). Such tax incentives may work to attract more crowd investors. Thus, we test the following hypothesis:

H4

Campaigns launched by firms eligible for the Angel Tax System are more likely to succeed in equity crowdfunding than others.

Finally, we examine the relationship between the campaign target amount and campaign success. Setting an unrealistically high goal may discourage crowd investors from participating (Ralcheva & Roosenboom, 2020). Campaigns with a high target amount need to attract more crowd investors because of a limited amount of investment per firm each year; specifically, an investor in Japanese equity crowdfunding is permitted to invest no more than 500 thousand JPY per firm each year. Especially on equity crowdfunding platforms with the all-or-nothing rule, crowd investors pay more attention to the probability of campaign success. The probability of a lower target amount is higher than that of a higher one. Indeed, previous studies have found that campaigns with a lower target amount have a higher probability of campaign success (Cumming et al., 2021; Piva & Rossi-Lamastra, 2018; Ralcheva & Roosenboom, 2020; Vulkan et al., 2016). We test the following hypothesis:

H5

Campaigns with a lower target amount are more likely to succeed in equity crowdfunding than others.

4 Data and methods

4.1 Data source

We construct a sample of equity crowdfunding campaigns listed on FUNDINNO, owned by FUNDINNO, Inc. (formerly Japan Cloud Capital, Inc.), an equity crowdfunding platform service initially founded in Japan. FUNDINNO is a leading equity crowdfunding platform, and as already mentioned, approximately 78% of equity crowdfunding campaigns in Japan were listed on FUNDINNO as of August 2021.

FUNDINNO can be regarded as an entrepreneur-led crowdfunding platform. Entrepreneurs who require equity finance apply their campaigns to the FUNDINNO platform. FUNDINNO conducts strict screening of qualifications based on: (i) issuers and their existence, (ii) financial conditions, (iii) business plans, (iv) eligibility, (v) management soundness, (vi) corporate structure for business continuity, (vii) risks associated with investment in the issuers’ over-the-counter securities, (viii) corporate governance, (ix) usage of funds raised, (x) validity of the campaign target amount, (xi) business reports and statements according to the laws, and (xii) ability for periodic reports to shareholders and holders of stock options. After the review process by FUNDINNO, entrepreneurs can launch their campaigns on the platform, and FUNDINNO provides due diligence throughout the process. FUNDINNO offers two investment methods: common shares and stock options. While common shares are linked to voting rights, stock options are not. FUNDINNO allows individuals (i.e., crowd investors) to participate and invest in campaigns listed on the platform—more typically, with an investment amount of 100 thousand to 500 thousand JPY—through user registration.¹⁵

FUNDINNO follows the all-or-nothing rule. On this platform, equity crowdfunding campaigns involve 10 types of information: (i) VC backing, (ii) corporate venture capital (CVC) backing, (iii) angel investor backing, (iv) past funding experience in FUNDINNO, (v) profitability in the preceding accounting year, (vi) existence of developed services and products, (vii) patent applications/registrations, (viii) common shares or stock options, (ix) eligibility for the Angel Tax System, and (x) shareholders’ benefits. This information would be useful to crowd investors when they make investment decisions.¹⁶

4.2 Sample

The sample consists of 242 campaigns conducted between November 2017 and August 2021.¹⁷ In the sample, 186 campaigns were successful, with the proportion of successful campaigns accounting for approximately 77%. As shown in Table 1, this average success, specifically the mean probability of campaign success, is much higher than that in other economies, except for German equity crowdfunding. This is partially due to the rigorous screening process by equity crowdfunding platforms and the limitation of the upper bound of equity finance to less than 100 million JPY in Japan.

Figure 2 shows the distribution of the ratio of the amount pledged in the sample. The ratio is less than one if the campaign fails and not less than one if the campaign succeeds. Many successful campaigns received twice the target amount, and some received approximately four times or more of the target amount. In contrast, many unsuccessful campaigns failed by a wide margin, while unsuccessful campaigns failed by a small margin is rare. These findings indicate that the amount pledged on the equity crowdfunding platform is highly skewed, which is consistent with Agrawal et al. (2014).

Fig. 2

Distribution of the ratio of the amount pledged to the campaign target amount. Note The total number of campaigns is 242. Amount pledged/Target amount indicates the ratio of the amount pledged to the campaign target amount. When a campaign fails, the ratio is less than one

4.3 Methods

Most empirical studies have estimated the determinants of campaign success using binary response models, such as logit and probit models (Piva & Rossi-Lamastra, 2018; Ralcheva & Roosenboom, 2020). We use a binary response model—specifically, a probit model—to identify the factors affecting the probability of campaign success.

We also employ a survival analysis approach to consider the differences in the time to campaign success between campaigns. While 77% of the 242 campaigns were successful, some succeeded soon after their launch. We exhibit the survival estimator proposed by Kaplan and Meier (1958) and estimate the proportional hazards model proposed by Cox (1972), following the literature (Felipe et al., 2022).¹⁸ In this study, the Kaplan–Meier (K-M) survival estimator is used to capture the proportion of campaigns that have not yet achieved the target amount at time t. The proportional hazards model is also used to identify the factors affecting the time to campaign success. By focusing on the time to campaign success, we provide evidence of how fast campaigns succeed in equity crowdfunding.

More importantly, we include the campaign target amount in the binary response model of campaign success and identify whether the campaign target amount affects the probability of campaign success. Firms that launch campaigns together with the platform firm (FUNDINNO, Inc.) decide the campaign target amount in advance, which may vary across campaigns or firms. Therefore, we estimate the determinants of the campaign target amount in the regression model. Such an estimation of the campaign target amount helps us understand how the factors affect the probability of campaign success, that is, whether they affect the campaign target amount rather than its probability. Furthermore, for robustness, we employ the probit model with a continuous endogenous covariate because the campaign target amount may be appropriately treated as endogenous. To employ this model, we estimate the equation of the campaign target amount with instruments, in addition to the covariates used in the equation of campaign success.

4.4 Variables

Campaign success is measured by a binary variable.¹⁹ We calculate the cumulative amount pledged by the campaign to measure this variable.²⁰ Campaign success (SUCCESS), the dependent variable in this study, takes a value of one if the (cumulative) amount pledged in equity crowdfunding is not less than the campaign target amount. We also employ the time to campaign success, measured by hour, to employ the survival analysis approach.

As discussed, the probability of campaign success may depend on the campaign target amount. For the determinants of campaign success, we measure the target amount (TARGET), defined as the logarithm of the campaign target amount (million JPY). As already explained, we also use the campaign target amount as the dependent variable and estimate its determinants.

Regarding the variables for hypothesis testing, we measure patents (PATENT) using a dummy variable that takes a value of one if the firm gains or applies for a patent. We use a dummy variable for services and products (SERVICE) that takes a value of one if the firm has already provided services and products prior to equity crowdfunding. Regarding professional investors, VC backing (VC) is measured by a dummy variable that takes a value of one if the firm is backed by VC prior to equity crowdfunding. In addition to VC backing, CVC backing (CVC) is measured by a dummy variable for the firm backed by a non-financial firm or CVC. We also use a dummy variable for angel investors (ANGEL) for the firm backed by angel investors. The Angel Tax System (ATS) is measured by a dummy variable that takes a value of one if the firm is eligible for the Angel Tax System, including both Types A and B.

As controls, we include paid-in capital, i.e., capital stock (CAPITAL), to control for the differences in capital size between campaigns. We also use team size (TEAM) to control for differences between campaigns, such as networks, according to the number of team members. In addition, we measure the firm’s funding experience in FUNDINNO before the launch of the campaign (EXPERIENCE) using a dummy variable to control for follow-on funding. Moreover, we measure firm age (AGE) and campaign start year (year dummies).

Furthermore, to identify the campaign target amount, we add the following instruments: campaign period (DURATION) and the number of crowd investors (CROWD) in FUNDINNO. These instruments are considered to be related to the campaign target amount because entrepreneurs set the campaign target amount based on the campaign period and crowdfunding market conditions. Table 7 in the Appendix provides the definitions of the variables used in this study and Table 8 presents the correlation matrix of the covariates.

4.5 Descriptive statistics

Table 2 presents descriptive statistics of the variables used in the regression equations. The mean amount pledged is approximately 28 million JPY, and the maximum amount pledged is 99 million JPY owing to the upper limit of equity financing, 100 million JPY. The mean target amount of campaigns is approximately 16 million JPY.

Table 2 Descriptive statistics of variables

Table 3 presents the means of the variables used, according to the success or failure of campaigns. While the mean amount pledged in the successful campaign group is larger than that in the failed campaign group, the mean target amount in the successful campaign group is smaller than that in the failed campaign group. Campaigns that seek to raise larger funds from crowd investors do not succeed in equity crowdfunding. Moreover, VC and ATS differ between successful and failed campaigns, indicating that campaigns launched by VC-backed firms and those by firms eligible for the Angel Tax System are more likely to succeed in equity crowdfunding. However, PATENT, SERVICE, CVC, and ANGEL do not significantly differ between successful and failed campaigns.

Table 3 Mean differences between successful and failed campaigns

5 Results

5.1 Campaign target amount

Table 4 presents the estimation results for the campaign target amount (TARGET) when the ordinary least squares method is employed. Column (i) in Table 4 reports the estimation results when we regress the campaign target amount on the covariates that will be used in the equation of campaign success, in addition to the two instruments (DURATION and CROWD). However, as shown in Table 8, the Angel Tax System (ATS) is negatively correlated with firm age (AGE) because firms under 10 years of age are eligible for the Angel Tax System; thus, firm age (AGE) is excluded from column (iii). In addition, capital size (CAPITAL) tends to be positively correlated with private equity capital (i.e., VC, CVC, and angel investors); thus, capital size (CAPITAL) is also excluded from columns (ii) and (iii) to avoid multicollinearity.

Table 4 Estimation results for the campaign target amount

As shown in Table 4, the patent dummy (PATENT) has a positive and significant effect on the campaign target amount. This indicates that firms with intellectual capital are more likely to set a higher target amount, presumably because they require more funding for R&D. In contrast, the services/products dummy (SERVICE) has a negative and significant effect on the campaign target amount, indicating that campaigns that have already provided services and products are more likely to have a lower target amount. Previous studies have argued that equity crowdfunding is used for marketing purposes (Mollick, 2014). Several firms may aim to investigate the market and advertise their services and products rather than raise funds on equity crowdfunding platforms. Moreover, the number of crowd investors (CROWD) has a positive and significant effect on the campaign target amount. As shown in Fig. 1, equity crowdfunding has recently been developed in Japan. As crowd investors on equity crowdfunding platforms increase, entrepreneurs may expect to raise more funds in equity crowdfunding.

5.2 Campaign success

Table 5 presents the estimation results for campaign success (SUCCESS) when the probit model is employed, and a positive coefficient indicates that the factor increases the probability of campaign success. Column (i) in Table 5 reports the estimation results when the above covariates (PATENT, SERVICE, VC, CVC, ANGEL, ATS, CAPITAL, TEAM, EXPERIENCE, AGE, TARGET, and year dummies) are included in the regression equation. However, as previously mentioned, capital size (CAPITAL) and firm age (AGE) are correlated with the main covariates; thus, these covariates are excluded from columns (ii), (iv), and (vi). In addition, the campaign target amount (TARGET) may be endogenously determined; thus, it is excluded from columns (iii) and (iv). Moreover, the regression equation for the campaign target amount (TARGET) is estimated in Table 4, and the campaign target amount may be endogenously determined. Thus, we present the estimation results in columns (v) and (vi), using the probit model with a continuous endogenous covariate. To identify the campaign target amount, we add DURATION and CROWD as instruments. However, the Wald test of the exogeneity of the instrumented covariates does not reject the null hypothesis of no endogeneity; consequently, the results assuming their exogeneity are preferable.²¹ Overall, the results presented in columns (v) and (vi) are consistent with those in the other columns; hence, the results are robust, irrespective of the endogeneity of the campaign target amount.

Table 5 Estimation results for the probability of campaign success: Probit model (coefficients)

As Table 5 shows, overall the patent dummy (PATENT) has a positive effect on the probability of campaign success. The results indicate that campaigns launched by firms with patent applications and registrations are more likely to succeed in equity crowdfunding than others, which supports H1. The results indicate that patenting positively affects campaign success as a signal of project quality to achieve campaign success. While Rossi et al. (2021) did not find a significant effect on campaign success regarding entrepreneur-led platforms in the UK and US, our results show a positive and significant effect regarding entrepreneur-led platforms in Japan. One reason is that entrepreneurs pursuing technological development may seek entrepreneur-led platforms, partly because Japan has underdeveloped private equity capital and entrepreneurs lack options for investor-led platforms. In addition, the upper amount of funding is limited even for technology-based campaigns that require more funds. Such campaigns may attract more crowd investors because of their high ability and innovative potential; hence, they have a higher probability of campaign success.

Meanwhile, the coefficients of patents are insignificant in columns (iii) and (iv) when the campaign target amount is uncontrolled. In other words, as shown in columns (i), (ii), (v), and (vi), we find that the probability of campaign success is higher for firms with patent applications and registrations when controlling the campaign target amount. As shown in Table 4, the patent dummy is positively associated with the campaign target amount; thus, the target amount differs according to patent applications and registrations. While innovative firms set a higher target amount, campaigns launched by such firms are more likely to attract crowd investors because of their high ability and innovative potential. Our findings also imply that the threshold level of campaign success differs among campaigns, depending on the types of campaigns, such as technology-based ones.

The services/products dummy (SERVICE) has a negative effect on the probability of campaign success, although it is insignificant in columns (i), (iii), and (iv). While the results indicate that campaigns associated with the development of services and products have a lower probability of campaign success, our results do not fully support H2. As Table 4 shows, the services/products dummy is negatively associated with the campaign target amount. These findings suggest that campaigns associated with the development of services and products do not exhibit a higher probability of campaign success, even though firms set a lower target amount to increase the probability of campaign success.

VC backing (VC) has a positive and significant effect on the probability of campaign success. The results indicate that campaigns launched by VC-backed firms are more likely to succeed in equity crowdfunding than others. We provide support for H3. This indicates that VC, other than CVC and angel investors, attracts more crowd investors, partly because of VC’s experience in selecting investments with due diligence. However, we find no significant relationships regarding CVC and angel investor backing. In other words, campaigns launched by firms backed by CVC, including non-financial firms, and those backed by angel investors do not necessarily have a higher probability of campaign success. This also suggests that the impact of professional investors on the probability of campaign success varies according to the type of investor.²²

The Angel Tax System dummy (ATS) has a positive and significant effect on the probability of campaign success, although insignificant, the endogeneity of the campaign target amount is considered in columns (v) and (vi). The results indicate that campaigns launched by firms eligible for the Angel Tax System are more likely to succeed in equity crowdfunding than others, almost supporting H4. This may also suggest that crowd investors tend to crowd such campaigns.

As for the controls, the coefficients of the number of team members (TEAM) are positive and significant for campaign success. The results indicate that campaigns launched by large teams are more likely to succeed in equity crowdfunding, suggesting that such campaigns have advantages for investor networks.

Furthermore, the campaign target amount (TARGET) has a negative and significant effect on the probability of campaign success. A negative relationship is found even when the endogeneity of the campaign target amount is considered in columns (v) and (vi). We provide support for H5. The results indicate that firms that set to a lower target amount have a higher probability of campaign success, consistent with previous studies (Cumming et al., 2021; Piva & Rossi-Lamastra, 2018; Ralcheva & Roosenboom, 2020; Vulkan et al., 2016).

5.3 Time to campaign success: survival analysis approach

To demonstrate how fast campaigns succeed, we calculate the K-M survival estimates according to private equity participation prior to equity crowdfunding. Figure 3 displays the K-M survival estimates when the sample is divided according to the following dummies: (i) PATENT, (ii) SERVICE, and (iii) ATS. These figures indicate that the proportion of campaigns that have not achieved the target amount changes over time from the launch. Figure 4 also displays the K-M survival estimates when the sample is divided according to the following dummies: (i) VC, (ii) CVC, and (iii) ANGEL.

Fig. 3

Kaplan–Meier survival estimates: Patent, services/products, and the Angel Tax System. Note The figures illustrate the Kaplan–Meier survival estimates (KM survival estimates) of campaign success between the two groups. The total number of observations is 242. Chi-square log-rank test statistics are 4.40 (p < 0.05), 0.01 (p > 0.1), and 6.20 (p < 0.05) for patent, services/products, and the Angel Tax System, respectively

Fig. 4

Kaplan–Meier survival estimates: VC, CVC, and angel investor backing. Note The figures illustrate the Kaplan–Meier survival estimates (KM survival estimates) of campaign success between the two groups. The total number of observations is 242. Chi-square log-rank test statistics are 7.62 (p < 0.01), 0.77 (p > 0.1), and 0.22 (p > 0.1) for VC, CVC, and angel investor backing, respectively

As shown in Figs. 3 and 4, many campaigns achieve the target amount within 24 h of the launch. In Fig. 3, approximately 78% of campaigns launched by firms with patent applications and registrations achieve the target amount within 24 h. Moreover, approximately 80% of campaigns launched by firms eligible for the Angel Tax System have achieved this. We also find significant differences in the K-M survival estimates for patents and the Angel Tax System. In contrast, there is no difference in the development of services and products. The findings using K-M survival estimates support H1 and H4. In Fig. 4, approximately 82% of campaigns launched by VC-backed firms achieve the target amount within 24 h. We find a significant difference in the K-M survival estimates between VC backing and others. In contrast, there are no differences between CVC-backed firms and others, and between angel investor-backed firms and others. The findings on the impact of VC backing on campaign success are consistent with those in Table 5.

Table 6 presents the estimation results for the time to campaign success when the proportional hazards model is employed. A positive coefficient indicates that the factor accelerates the time to campaign success. In Table 6, we use the covariates included in columns (i)–(iv) of Table 5.

Table 6 Estimation results for the time to campaign success: Proportional hazards model (coefficients)

The patent dummy has a positive and significant effect on the time to campaign success. We find that patenting is positively associated with the time to campaign success, as well as the probability of campaign success. The results indicate that campaigns launched by firms with patent applications and registrations are more likely to succeed in equity crowdfunding earlier than others.

However, the services/products dummy has no significant effect on the time to campaign success in Table 6. We do not find any evidence of the impact of the development of services and products on the time to campaign success. Although campaigns that develop their services and products tend to maintain a certain level of campaign success by setting a lower target amount, they do not necessarily achieve the campaign target amount faster in equity crowdfunding.

The VC dummy has a positive effect on the time to campaign success at the 1% level in Table 6, indicating that campaigns launched by VC-backed firms are more likely to succeed in equity crowdfunding earlier than others. As shown in Tables 5 and 6, the results reveal that VC participation prior to equity crowdfunding positively affects the time to campaign success. Similarly, we find no evidence of the impact of CVC and angel investor backing on the time to campaign success.

Furthermore, the Angel Tax System dummy has a positive effect on the time to campaign success at the 1% level. The findings on the time to campaign success are consistent with those on the probability of campaign success. The results reveal that campaigns launched by firms eligible for the Angel Tax System are more likely to succeed in equity crowdfunding earlier than others, suggesting that the Angel Tax System effectively attracts crowd investors.

5.4 Discussions

Numerous scholars have addressed equity crowdfunding since the 2010s, and the majority of samples used in the literature on crowdfunding tend to be primarily focused on European economies and the US. Only a few studies have included data from other economies, such as Australia, Brazil, Canada, China, India, New Zealand, Singapore, and Taiwan (Mochkabadi & Volkmann, 2020). To the best of our knowledge, there has been no academic empirical research on equity crowdfunding in Japan, where private equity capital, including informal investors, has not yet been developed (Honjo & Nakamura, 2020). Under equity crowdfunding with a limited amount of funding, we provide new findings on the behavior of entrepreneurs and crowd investors. Although the results using the case of campaigns in Japan may stem from the country’s economic conditions, the findings are useful in gaining insights into policies that can develop financial alternatives to promote entrepreneurial innovation in other economies that depend on bank financing.

This study provides some implications about equity crowdfunding by examining the factors that affect campaign success and the target amount. Some studies have examined public support, such as tax incentives, for investments via equity crowdfunding. Still, the results did not reveal a significant effect of tax incentives on the probability of campaign success (Vismara, 2016, 2018a, 2019). We provide significant evidence of the impact of policy support on campaign success regarding Japan’s Angel Tax System. Our findings suggest that the Angel Tax System help raise funds on equity crowdfunding platforms for new businesses. Policymakers should consider how to design effective tax incentives for entrepreneurial finance to promote risky investment by crowd investors. Although we provide evidence only on the impact of the Angel Tax System on the probability of campaign success, further studies on policy support would help to promote new firm creation and start-up performance via equity crowdfunding.

Moreover, we provide novel insights into the mechanisms that affect the probability of campaign success by including the campaign target amount. While innovative firms, measured by patenting, tend to set a higher target amount, they also have a higher probability of campaign success. Even if innovative firms require a high target amount for technology-based campaigns, they are more attractive to crowd investors. Although individual investment is limited to a small amount (i.e., no more than 500 thousand JPY), crowd investors may find it easier to participate in such small-sized investments. In addition, while campaigns that have developed their services and products tend to have a lower target amount, they do not succeed in equity crowdfunding. This implies that they seek to promote their services and products rather than raise funds on equity crowdfunding platforms. In this respect, equity crowdfunding platforms are required to enhance due diligence services because due diligence leads to higher funds raised and a greater percentage of successful campaigns (Rossi & Vismara, 2018). Furthermore, such interpretation arises from the findings of the relationships with the campaign target amount. In this respect, our empirical approach to estimating the campaign target amount, following previous studies (Rossi et al, 2021), is useful in better understanding how the campaign succeeds in equity crowdfunding.

Not surprisingly, new financing techniques, including equity crowdfunding, are often regulated, dependent on the country context. More importantly, the findings on equity crowdfunding, especially for the promotion of services and products, may be caused by limitations in equity finance—specifically, 100 million JPY per year on Japanese equity crowdfunding platforms—which is lower than in other economies, such as the UK and the US. While the limitation to the small amount of funding leads to a higher probability of campaign success, it may result in distorted purposes, such as promoting services and products, which may stem from stringent regulations due to strong investor protection in equity crowdfunding. For policymakers and crowdfunding platform firms, investor protection against adverse selection is essential to ensure the sustainability of equity crowdfunding markets (Walthoff-Borm et al., 2018a, 2018b). Importantly, unsophisticated investors require more protection through regulations (Goethner et al., 2021). However, overly strong investor protection, particularly relevant to equity crowdfunding in Japan, may harm small firms, including entrepreneurial initiatives. This contrasts with the traditional law and finance view that stronger investor protection is better (Hornuf & Schwienbacher, 2017). Our findings based on strictly regulated equity crowdfunding platforms could be useful in considering effective regulations for equity crowdfunding.²³

However, this study has some limitations. First, we did not examine post-campaign outcomes because of the lack of data on firm performance, although some studies have examined them (Eldridge et al., 2021; Hornuf et al., 2018, 2022; Signori & Vismara, 2018). Therefore, whether firms can improve their performance by raising funds on equity crowdfunding platforms is unclear. Second, we did not consider alternative financing instruments like subsidies and public loan programs. Further research is required to evaluate the role of equity crowdfunding in nurturing start-up firms with growth potential.

6 Conclusions

This study investigated campaign success in equity crowdfunding, using campaigns listed on a leading Japanese equity crowdfunding platform. We found that patenting positively affects campaign success. The results indicate that innovative firms have a higher probability of campaign success, suggesting that equity crowdfunding has the potential to help finance R&D activities. We also provided evidence that campaigns that have already provided services and products have a lower probability of success, although this relationship is not always significant. Such firms tend to set a lower target amount, suggesting that they aim to investigate the market and advertise their services and products rather than raise funds on equity crowdfunding platforms. Moreover, campaigns launched by VC-backed firms are more likely to succeed than others. Furthermore, campaigns launched by firms eligible for the Angel Tax System have a higher tendency to succeed in equity crowdfunding.

Appendix

Table 7 summarizes the definitions of variables used in this study. Table 8 presents the correlation matrix for the covariates. Table 9 presents the estimation results for campaign success based on the duration data when the endogeneity of the campaign target amount is considered. In Table 9, we use a discrete-time duration model because a continuous endogenous covariate cannot be included in the proportional hazards model (i.e., the continuous-time duration model). To estimate the discrete-time duration model, we define a binary variable (SUCCESS_P), which equals one at time t when SUCCESS = 1 for the campaign. To consider campaign-specific effects, we use the cluster variance of the campaign as the cross-section identifier. To specify the time-specific term, we also include the logarithm of hours from the launch (HOUR). As shown in Table 9, the coefficients of PATENT, SERVICE, VC, and ATS are significant for the time to campaign success. However, the Wald test of the exogeneity of instrumented variables does not reject the null hypothesis of no endogeneity.

Table 7 Definitions of variables used in this study

Table 8 Correlation matrix

Table 9 Estimation results for the time to campaign success: Discrete-time duration model (probit model with a continuous endogenous covariate)